1.3 Development steps this week. #2

Development Step #2 / Tracking a face with Open CV and using USB Camera.

o Approach.

• Face Detection

For this development step I will be using the examples provided in the OpenCV library in Processing and test out the different tracking accuracy of facial features.

• USB Camera

I am looking for a way to capture video, which is streamed through an USB port. I am going to look for help through different forums.

The camera that I plan on using is Logitech c310.

o Learning Experience.

• Face Detection

First I started with face detection out of a static image.

Input:

Code: https://gist.github.com/TheVisualG/ada0a95f989c5c333a197a2d1a2bf75f

Output:

I Learned:
in order to activate the face detection in Processing we need to use:

Rectangle[] faces;
opencv.loadCascade(OpenCV.CASCADE _FRONTALFACE); // I can select different parts of the body as a detection target.

faces = opencv.detect();

• Live Video

Input:
Live video streamed from my laptop camera.

Code:

	import gab.opencv.*;
	import processing.video.*;
	import java.awt.*;
	Capture video;
	OpenCV opencv;
	void setup() {
	size(640, 480);
	video = new Capture(this, 640/2, 480/2);
	opencv = new OpenCV(this, 640/2, 480/2);
	opencv.loadCascade(OpenCV.CASCADE_FRONTALFACE);
	video.start();
	}
	void draw() {
	scale(2);
	opencv.loadImage(video);
	image(video, 0, 0 );
	noFill();
	stroke(0, 255, 0);
	strokeWeight(3);
	Rectangle[] faces = opencv.detect();
	println(faces.length);
	for (int i = 0; i < faces.length; i++) {
	println(faces[i].x + "," + faces[i].y);
	rect(faces[i].x, faces[i].y, faces[i].width, faces[i].height);
	}
	}
	void captureEvent(Capture c) {
	c.read();
	}

view raw

live_video

hosted with ❤ by GitHub

Output:

I learned:

Streaming live video in Processing using following commands:

import processing.video.*;
Capture video;
video = new Capture(this, 640/2, 480/2);
video.start(); image(video, 0, 0 );
void captureEvent(Capture c)
{ c.read();
}

• USB Camera

I googled around and bumped into this post from the Processing forums:
https://forum.processing.org/two/discussion/7736/using-usb-camera-instead-of- isight

	import processing.video.*;
	Capture cam;
	void setup() {
	size(1280, 960);
	String[] cameras = Capture.list();
	if (cameras == null) {
	println("Failed to retrieve the list of available cameras, will try the default…");
	cam = new Capture(this, 640, 480);
	} if (cameras.length == 0) {
	println("There are no cameras available for capture.");
	exit();
	} else {
	println("Available cameras:");
	printArray(cameras);
	// The camera can be initialized directly using an element
	// from the array returned by list():
	cam = new Capture(this, cameras[80]);
	// Or, the settings can be defined based on the text in the list
	//cam = new Capture(this, 640, 480, "Built-in iSight", 30);
	// Start capturing the images from the camera
	cam.start();
	}
	}
	void draw() {
	if (cam.available() == true) {
	cam.read();
	}
	image(cam, 0, 0, width, height);
	// The following does the same as the above image() line, but
	// is faster when just drawing the image without any additional
	// resizing, transformations, or tint.
	//set(0, 0, cam);
	}

view raw

camera_check

hosted with ❤ by GitHub

o Result.

• Face Detection

Input:

Code:

	import gab.opencv.*;
	import java.awt.Rectangle;
	OpenCV opencv;
	Rectangle[] faces;
	void setup() {
	opencv = new OpenCV(this, "profile_test.jpg");
	size(960, 960);
	opencv.loadCascade(OpenCV.CASCADE_FRONTALFACE);
	faces = opencv.detect();
	}
	void draw() {
	image(opencv.getInput(), 0, 0);
	noFill();
	stroke(0, 255, 0);
	strokeWeight(3);
	for (int i = 0; i < faces.length; i++) {
	rect(faces[i].x, faces[i].y, faces[i].width, faces[i].height);
	}
	}

view raw

face_detection_test

hosted with ❤ by GitHub

Output:

When it comes to tracking nose, eye, mouth there are many false-positives. That is why I believe frontal face and profile are more optimal options, or more detailed tracking can be used with pain background, without any texture that might bring an error.

opencv.loadCascade(OpenCV.CASCADE _FRONTALFACE);
opencv.loadCascade(OpenCV.CASCADE _PROFILE);
opencv.loadCascade(OpenCV.CASCADE _EYE);
opencv.loadCascade(OpenCV.CASCADE _MOUTH);

 

• Live Video

Output:

 

• USB Camera.

I plugged the USB camera and ran the code, but unfortunately the only thing that was detected was my laptop camera.

I tried different USB ports, checked if the camera runs on other computers and it did. A􀁛er long struggles I ended
up manually reinstalling the UBS driver from the official website:

http://support.logitech.com/en_us/product/hdwebcam-c310/downloads

Finally I succeeded and I had the full list of camera optioons (resolution and fps) :

Now in order to switch between camera I need to change the number in this line of code:

cam = new Capture(this, cameras[80]);

Code:

	import processing.video.*;
	Capture cam;
	void setup() {
	size(1280, 960);
	String[] cameras = Capture.list();
	if (cameras == null) {
	println("Failed to retrieve the list of available cameras, will try the default…");
	cam = new Capture(this, 640, 480);
	} if (cameras.length == 0) {
	println("There are no cameras available for capture.");
	exit();
	} else {
	println("Available cameras:");
	printArray(cameras);
	cam = new Capture(this, cameras[80]);
	cam.start();
	}
	}
	void draw() {
	if (cam.available() == true) {
	cam.read();
	}
	image(cam, 0, 0, width, height);
	}

view raw

cam_check_2

hosted with ❤ by GitHub

o Sources.

My sources are from the Processing examples included in the OpenCV library.
https://forum.processing.org/two/discussion/7736/using-usb-camera-instead-of- isight
http://support.logitech.com/en_us/product/hdwebcam-c310/downloads